AMD rumors

DoomArse · Aug 18, 2003

what do the model numbers on AMD's mean, I've always heard that they stand for how they compare to pentiums (a 1700+ is comparable to a 1700 mhz pentium) however this dosent make sense to me. Can someone please explain this.

Bruno · Aug 18, 2003

Originally it was to compare to the old Thunderbird Athlons. But, we all know that's BS and it truly is to play a numbers game w/ intel in comparing to P4's.

Gautam · Aug 18, 2003

What the PR ratings actually are are how they Athlon XP's compare to the original Athlon. They are based off a T-Bird 1.4ghz. Their obvious intent is to compare the Athlon XP to Pentium 4's, however, officially, they do not.

c627627 · Aug 18, 2003

Question is, what the heck is Athlon 64 FX-51, 53, 55?

Why 5x?

http://www.pbase.com/image/17079307/original

P.S. By the way, off topic, I don't see Athlon XP 3400 or 3600 happening? What do you guys think?

Gautam · Aug 18, 2003

Why 5x?

5's a cool number?

P.S. By the way, off topic, I don't see Athlon XP 3400 or 3600 happening? What do you guys think?

Very, very likely, IMHO. The Clawhammers clearly aren't going to deliver, at least not in the initial stages. AMD's going to need something to really challenge the seemingly problematic Prescotts. People are going to want to hang on to their Socket A configs for a while longer, also. Not to mention how wonderfully the new JIXHB's are overclocking. I'd say that a 2.4ghz default Barton should be very easy for AMD to fab now, or at least in the near future.

c627627 · Aug 18, 2003

Gautam said:
5's a cool number?

What about this: ... although Prescott may not translate into a new official core name, Intel Tejas due out Q4 2004 most probably will.

And if AMD thinks 5 comes after 4 (Pentium 4)... why not get a head start?

Maybe not.

Col_Loki · Aug 18, 2003

Personally i dont think the 64bit systems are going to keep AMD afloat without developing there Xp line aswell as its there bulk sales that are bringing in the money.

Granted they have the total monopaly of the 64bit market i think it would be stupid not to compete with intel in the other. We now know AMD are bringing out the thorton (based on the barton) which makes me think they arent phasing out the production line for quite a bit yet.

Theres my 2c worth, very interested to see what ye guys are thinkin - im definetly expectin an Xp3600+ atleast.

c627627 · Aug 18, 2003

Thorton's going to be a Duron they're now saying...

Athlon XP 3600+ would be a 2.4 GHz chip. It would really have little or no headroom for overclocking at all.

Have we seen many chips do more than 2.4 GHz with all default settings?

You know highest T-Bred B was the 2800+ at 2.25 with a headroom of about 100 MHz or so...

2.4 would need over 2.55 GHz headroom, wouldn't it?

Gautam · Aug 18, 2003

How about 2.3ghz at only 1.5v, and 2.7ghz or air? Linkage. Like I've said before, T-Breds are going to get progressively better. They haven't yet hit the wall for the .13 micron process. Getting there, but not quite. I'd expect 3600+'s to have at least 300mhz of headroom with decent cooling.

c627627 · Aug 18, 2003

Regardless of what they actually release based on marketing considerations, what do you see as the wall for Barton commercially?

(2.4 retail would need over 2.55 GHz headroom, wouldn't it?)

Gautam · Aug 18, 2003

My prediction is that the end will probably be a 3800+, stock voltage 1.5v-1.6v, stock speed 2500mhz. AMD's stock cooling may get slightly upgraded, but not drastically. With the high-end air cooling in the hands of most on these forums, I'd expect 500mhz headroom, but with low amounts of voltages. The last T-Breds aren't going to sustain stable overclocks at much higher than 1.75v, and this is the max that I'd expect on air. 2.9-3.0ghz, with 1.75v-1.8v.

c627627 · Aug 18, 2003

3 GHz would be Athlon XP 5000+.

That's a bold statement and time will tell.

Gautam · Aug 18, 2003

Yes, but the official limit will only be at around 2.5ghz. It will be dangerous(by stock standards) to go much higher. And the 5000+ rating is from the dated formula. I'd say around a 4200+ equivalent. I'm basing my predictions partly on this:

Originally posted by hitechjb1
Its transistors have lower threshold characteristics due to process variation which produces transistors with shorter channel length. Shorter channel means lower transistor threshold, runs faster, draws larger leakage current and higher active current (hence higher active power). According to AMD spec, the 1.5V 1700+ has higher rated current than the 1.6V 1700+ (about 7% more).

Threshold voltage of a transistor is the gate voltage above which the transistor will conduct current orders of magnitude higher from source to drain compared to that below the threshold. Chips with lower threshold transistors can perform equally well with a lower supply voltage (Vcore) as those with higher threshold, because the transistors can conduct at a lower gate voltage.

This is normal for a given silicon process (say 0.13u) to have such variation that some transistors in certain chip die have shorter channel length (less than 0.13u) or some have longer channel length. Those that have shorter channel length have faster intrinsic speed and can run as fast when smaller Vcore is applied (pros). On the other hand (cons), due to the lower threshold voltage which draws higher leakage current and generates more heat at the same higher Vcore, these chips can run as fast at a low Vcore as the higher Vcore rate chips, but they will max out at a lower Vcore compared to the higher Vcore rated siblings.

The 1700+ has a run-away current at a lower Vcore compared to the 2100+. Run-away current refers to the leakage current and the heat generated positively feeding each other resulting instability.

The channel length will probably be significantly shorter with the last T-Breds. This is what is going to allow them to run so much faster, but also make higher levels of overclocking significantly detrimental to the processor.

c627627 · Aug 18, 2003

Thank you for that.

Here's the marketing perspective: 3.4 GHz Pentium 4 is going to be here this year so 3400+ should be a Q4 release, probably not October but November-December 2003.

However 3.6 GHz Pentium 4 is not expected until well into 2004.
Since 3.8 GHz Pentium 4 is thought to be a mid 2004 release...

you see Athlon XP's being manufactured at the end of 2004 and beyond?

DoomArse said:
what do the model numbers on AMD's mean

http://www.pbase.com/image/17079307/original
These woud be the changes after reading what Gautam said if the server weren't down:

2.3 GHz Athlon XP 3400+ (Q4 2003)
2.4 GHz Athlon XP 3600+ (Q1 2004)
2.5 GHz Athlon XP 3800+ (Q2 2004)

Gautam · Aug 18, 2003

Whoa, don't add any of my wild theories to your roadmap. Those are all guesses(accurate, I think, but who knows). I think that the Athlon XP's will be manufactured until just after the release of the San Diego. This way, the Clawhammers could become Durons, and the Socket A fully phased out. But the lifetime of the Athlon XP depends completely on the market. If people don't want to give up their Socket A mobos(not very likely, considering most seem to upgrade mobos on a yearly basis anyways) then the Athlon XP may last until mid 2004 at the latest or so. I think that AMD is going to try to beat out Intel and release a higher- PR rated processor than them in advance, so that they'll have a 3600+ well before Intel releases a 3.6ghz Prescott. The former should be far more feasable as well. Btw, I'm bookmarking this thread, just to see how correct I am a few months down the road

.

AMD rumors

DoomArse

Member

Bruno

Member

Gautam

Senior Benchmark Addict

c627627

c(n*199780) Senior Member

Gautam

Senior Benchmark Addict

c627627

c(n*199780) Senior Member

Col_Loki

Member

c627627

c(n*199780) Senior Member

Gautam

Senior Benchmark Addict

c627627

c(n*199780) Senior Member

Gautam

Senior Benchmark Addict

c627627

c(n*199780) Senior Member

Gautam

Senior Benchmark Addict

c627627

c(n*199780) Senior Member

Gautam

Senior Benchmark Addict

Similar threads